Acetaminophen toxicity and its prevention

 A recent analysis found that an FDA mandate limiting acetaminophen co-prescription with opiates did lead to impressive decreases in acetaminophen hepatotoxicity, though there was no significant difference in hepatotoxicity from acetaminophen by itself despite a black box warning and patient education (see opiates dec combo with acetamin dec liver failue JAMA2023 in dropbox, or doi:10.1001/jama.2023.1080)  

 

Details: 

-- two databases were accessed: 

    -- National Inpatient Sample (NIS): the largest publicly available all-payer US inpatient database with discharge data on more than 7 million inpatient stays annually 

    -- Acute Liver Failure Study Group (ALFSG): a prospective, 32 center cohort of adult patients with acute liver failure (ALF), defined by an INR >1.5, hepatic encephalopathy, no prior chronic liver disease, and disease duration < 26 weeks 

--  the FDA issued a mandate to limit the amount of acetaminophen in prescription combinations meds (including with opioids) to 325 mg per tablet in January 2011 with manufacturer compliance required by March 2014 , and a black box warning for over-the-counter preparations that an overdose can result in liver transplant or death

 

-- NIS: 39,606 patients with both acetaminophen and opioid toxicity; 468,025 with acetaminophen toxicity alone 

    -- median age 42, 67% women, 80% white/8% Hispanic/8% Black 

    -- history of alcohol misuse in 21%/suicide attempt 9%/hepatitis C 4% 

        -- these above numbers were for those having combination of acetaminophen and opioids; those with acetaminophen toxicity alone were younger at age 30 and more likely to be Hispanic/Black; the other demographics  were similar 

-- ALFSG: 465 patients with acetaminophen and opioid toxicity; 737 with acetaminophen toxicity alone 

    -- median age 39, 85% women, 82% white/7% Black/6% Hispanic 

    -- history of substance misuse 43% 

        -- these numbers  were from acetaminophen and opioid toxicity together; the numbers for acetaminophen alone were similar, though the median age was 34, 69% women and 38% had a history of substance misuse 

 

-- 3 time periods were measured for acetaminophen toxicity with and without opiates: 

    -- predicted incidence of hospitalization one day prior to the FDA announcement in 2011 versus afterwards 

    -- before versus after the final date for manufacturer compliance (March 2014) 

    -- before the FDA announcement in 2011 versus after the final date of manufacturer compliance in 2014 

-- each acetaminophen overdose case was assessed for intentionality 

 

Results: 

-- NIS: first analysis: predicted hospitalization 1 day prior to the FDA announcement in 2011 versus afterwards (too many numbers, so I will leave off the confidence intervals... but esp for NIS where the confidence intervals were quite narrow) 

    -- for the combination of acetaminophen and opioid toxicity:  

            -- 12.2 cases/100,000 hospitalizations prior to the announcement; by 2019: 4.4/100,000, p<0.001 

            -- the odds of hospitalizations increased 11%/ year prior to the announcement and decreased 11% per year after the announcement 

            -- and, the odds of acute liver failure (ALF) cases decreased from 27.4% to 5.3%, p<0.001 

        -- before versus after the final date of manufacturer compliance: 

            -- 10.4 cases/100,000 hospitalizations prior to the announcement; by 2019: 4.2/100,000, p<0.001 

            -- the odds of hospitalizations increased 1%/ year prior to the announcement and decreased 13% per year after the announcement 

        -- from the ALFSC study, the predicted percentages of ALF cases involving acetaminophen and opioid toxicity decreased from 27.4% to 5.3%, p<0.001 

-- NIS for acetaminophen toxicity by itself: 

            -- 110.0 cases/100,000 hospitalizations prior to the announcement; by 2019: 101.2/100,000, p<0.001 

            -- the odds of hospitalizations increased 1%/ year prior to the announcement (not statistically significant), and increased a statistically significant 4% per year after the announcement 

        -- from the ALFSC study, the predicted percentages of ALF cases involving just acetaminophen toxicity increased from 24.9% to 37.8%, p<0.001 

 

-- NIS: second analysis: before versus after the final date for manufacturer compliance: 

        -- for the combination of acetaminophen and opioid toxicity: 

            -- 10.4 cases/100,000 hospitalizations prior to the announcement; by 2019: 4.2/100,000, p<0.001 

            -- the odds of hospitalizations increased 1%/ year prior to the mandate and decreased 13% per year after it 

            -- from the ALFSC study, the predicted percentages of ALF cases involving acetaminophen and opioid toxicity decreased from 22.9% to 7.6%, p<0.001 

        -- NIS for acetaminophen by itself: 

            -- 87.2 cases/100,000 hospitalizations prior to the opiate mandate, by 2019: 106.9/100,000, p<0.001 

            -- the odds of hospitalizations decreased 4%  year prior to the announcement and there was no change  after the announcement 

        -- from the ALFSC study, the predicted percentages of ALF cases involving just acetaminophen toxicity increased from 27.2% to 34.1%, p<0.001 

 

-- NIS: third analysis: before FDA announcement versus after the final date for manufacturer compliance: 

        -- for the combination of acetaminophen and opioid toxicity: 

            -- 12.2 cases/100,000 hospitalizations prior to the announcement;  by 2019: 4.2/100,000, p<0.001 

            -- the odds of hospitalizations increased 11%/ year prior to the mandate and decreased 13% per year after it 

        -- from the ALFSC study, the predicted percentages of ALF cases involving acetaminophen and opioid toxicity decreased from 27.4% to 7.6%, p<0.001 

        -- NIS for acetaminophen by itself: 

            -- 110.0 cases/100,000 hospitalizations prior to the opiate mandate, 106.9/100,000, not significant 

            -- the odds of hospitalizations increased 1%  year prior to the announcement and increased 4% after,p<0.001  

        -- from the ALFSC study, the predicted percentages of ALF cases involving just acetaminophen toxicity increased from 24.9% to 35.1%, p<0.001 

 

 

-- Disease severity: 

    -- NIS, across all three analyses: no significant differences in in-hospital mortality prior to and after the mandate 

    -- ALFSG: no significant differences in peak bilirubin levels, creatinine levels, INR, MELD or MELD-Na scores (MELD score is followed to assess urgency of transplant; MELD-Na reflects severity of chronic liver disease), or 21-day survival across all three analyses; and no significant differences in peak AST and ALT levels, except in the first analysis 

        -- there was no significant difference in liver transplant in those hospitalized with acetaminophen and opioid toxicity, however there was a statistically significant increase in transplants for acetaminophen toxicity alone after the mandate in all three analyses 

-- all of the above was found after adjusting for seasonality, and excluding those with hepatitis C and alcohol misuse listed as one of top five diagnostic codes 

 

Commentary: 

-- acetaminophen is consumed pretty regularly:    

    -- it is found in more than 150 over-the-counter preparations for a variety of illnesses and symptoms, with varying quantities of acetaminophen (typically noted in the fine print, with no interpretation of what the number means) 

    -- in the US more than 25 billion acetaminophen doses are sold annually, either alone or in combination with other formulations

-- acetaminophen has a well-known hepatotoxicity which is dose-dependent and tends to have a relatively narrow window from therapeutic to toxic 

-- at least 60,000 people in the US are hospitalized with acetaminophen toxicity annually, though around half of these cases of acute liver failure are intentional overdoses 

    -- a 2005 study had found that 43% of acetaminophen overdoses were in people using acetaminophen and opioid medications. Studies done prior to the mandate did find that 6% of adults given the opioid/acetaminophen combination were prescribed more than 4000 mg per day of acetaminophen 

-- the dose at which acetaminophen hepatotoxicity occurs is somewhat variable. 44% of healthy volunteers given 4000 mg a day have clinically significant increases in their ALT levels. People who have chronic hepatitis C, concurrent alcohol use, older age, poor nutritional status,and fasting are at increased risk of hepatotoxicity at lower, therapeutic doses. genetics also plays a role. so, for some people, significantly lower doses than 4000 mg/d can be harmful [3,000 m/d may be more reasonable; and that is harder to reach if patients take the 325mg vs 500mg sized pills]

    -- and those with non-alcoholic fatty liver disease (NAFLD) are at higher risk of acetaminophen toxicity (see https://pubmed.ncbi.nlm.nih.gov/24575957/ ), and the incidence of NAFLD is increasing (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7132013/ ), and NAFLD happens in non-overweight individuals (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8683915/#:~:text=NAFLD%20used%20to%20be%20connected,or%20nonobese%20NAFLD%20(10).  Perhaps we should be screening  people more broadly for NAFLD. ?routine LFTs  ?routine ultrasounds…..

and many clinicians are not routinely assessing for NAFLD (ie, perhaps we should at least be checking liver function tests more regularly; and

-- the public health burden of acetaminophen is quite extensive (see Trends in rates of acetaminophen-related adverse events in the United States - PMC (nih.gov) :

    -- an estimated 59,000 ED visits annually; 41% (24,358) were classified as unintentional overdoses 

    -- an estimated 38,000 hospitalizations for acetaminophen-related poisonings

    -- and about 100000 Poison Center calls per year 

 

-- This current study, with its very large number of numbers noted above, basically found a few things: 

    -- after the FDA mandate limiting acetaminophen combined with opioids, there was a quite dramatic decline in the yearly rate of hospitalizations and cases of acute liver failure involving acetaminophen and opioid toxicity 

        -- though, ALF severity did not significantly decrease (??why: perhaps because many had irreversible liver failure at that point or they were the ones who continued to use acetaminophen for whatever reason)

    -- BUT, there was no decrease in hospitalizations for ALF from acetaminophen alone through public health education or the black box warning, in fact with an increase over time 

 

-- studies of the perceptions in the general population of the dangers of acetaminophen confirm they are quite low, despite these public health initiatives for education 

    -- the perception is lowest for combination products that are over-the-counter, one study found that only 15% of patients in a general medicine clinic could identify commonly prescribed combination products that contain acetaminophen 

    -- another study found that only half of the people surveyed correctly identified acetaminophen-containing products 

    -- and a population-based Canadian study (Impact of acetaminophen product labelling changes in Canada on hospital admissions for accidental acetaminophen overdose: a population-based study | CMAJ ) assessed the effects of updated  labeling standards around acetaminophen both in October 2009 and in 2016: they did not find any change in hospital admissions for accidental acetaminophen overdoses.  These labeling requirements focused initially on warnings about the potentially fatal risk of liver injury and the updated version in 2016 helped facilitate product identification and communicate safe dosing. The conclusion: modification of product labels did not reduce the rate of acetaminophen-related harm 

 

Limitations: 

-- this was a large database-mining assessment of two observational databases, the huge NIS one that had limited granular data, and the significantly smaller ALFSG one with much more granular data 

-- even the granular data we have does not include information about many important specifics: there is information on hepatitis C virus (though without information about the extent of liver damage associated with it or baseline liver function abnormalities) or alcohol use (the extent of alcohol consumption, past and current, as well as assessment of liver affects of the alcohol) 

-- these databases also are limited by the extent to which accurate data are meticulously recorded by individual clinicians or institutions (these data-mining studies require accurate ICD codes, coding for example acute liver failure along with coding opioids and/or acetaminophen). And, this was more likely to be a greater issue prior to the FDA mandate, when ALF was more commonly seen. Also, there is no specific code for intravenous drug use, so codes for heroin or methadone toxicity might have been used and these were not included in the data-mining for the study 

-- though the FDA mandate required compliance by March 2014, studies did show incomplete drug company adherence to that

 

So, this study and others mentioned above do seem to point to a few conclusions: 

-- though public health educational campaigns are important, they do not seem to have the effect of regulation and mandates. It is always an uphill battle for especially not so well-funded public health agencies to be able to compete with the massive advertising campaigns and self-serving practices of the drug companies (e.g. not mentioning dangers of the drugs, hiding ingredients in small print, etc.). the population overall seems to be inured to such warnings over time

    -- and, one could argue pretty strongly that the public health imperative is to decrease social harm from medications and other products, which really should include strong regulations to limit harmful products (e.g. severely limiting trans-fats seemed to work), eliminating or minimizing dangerous components of over-the-counter medications (e.g. eliminating acetaminophen in products where there is minimal evident benefit or decreasing acetaminophen dosaging when significant benefit can be documented) 

-- and this brings up the broader issue of pain management. There are a large number of nonmedical interventions that do help relieve pain, especially chronic pain.

    -- opioids, with many studies suggesting that they are no better than non-opioid meds  (eg see https://gmodestmedblogs.blogspot.com/2020/10/sciatica-early-pt-helps-longterm.html)  

    -- chronic pain management: the importance of non-medical therapies (eg CBT, minfulness, tai chi, PT, etc): https://gmodestmedblogs.blogspot.com/2017/04/home-based-cbt-for-low-back-pain.html

    -- But pretty much all the medications are fraught (acetaminophen as above, along with NSAIDs and opiates): 

        -- NSAIDs have their issues: an array of harms including hypertension, kidney disease, upper GI bleeds, heart failure, etc, though topical NSAIDs work well and are likely much safer (https://gmodestmedblogs.blogspot.com/2020/08/acute-musculoskeletal-pain-topical.html )

            -- a cautionary article suggesting that NSAIDs for acute pain may actually increase the likelihood of chronic pain: https://gmodestmedblogs.blogspot.com/2022/05/acute-pain-anti-inflammatories-lead-to.html

        -- and opiates are not only overprescribed but are often no more effective than non-opiates for both acute pain (kidney stones, post-op care) as well as chronic pain (low back pain, chronic knee pain). several of these studies have links in the recent blog on antidepressants for pain management: http://gmodestmedblogs.blogspot.com/2023/02/pain-management-review-of.html

    -- and this all means that we really do need better and safer pain management…

 

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs (2 options):

1. go to https://bucommunitymed.wpengine.com/, a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs and is the easiest to view blogs and displays more at a time.

2. go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list